R.F. OUTPUT POWER CONTROL

Description

BACKGROUND OF THE INVENTION

1. Technical Field

[0001] The present invention relates to the field of radio frequency (R.F.) output power control and more particularly to an R.F. power amplifier circuit comprising a power control loop including an R.F. power amplifier. The invention further relates to a method of controlling the output power of the R.F. power amplifier.

2. Description of the Prior Art

[0002] Modern R.F. applications like wireless communications devices require an efficient R.F. output power control for example to ensure a high transmission quality and to keep output signal fluctuations within limits that are defined in various standards.

[0003] Typically, R.F. output power control involves a power control loop including the R.F. power amplifier, a detector device for detecting the output power of the R.F. power amplifier and an error amplifier. Such an R.F. power control loop is known for example from US 5,378,996.

[0004] The power amplifier known from US 5,378,996 has a control input, a signal input, an R.F. power output and a detector output. The detector signal at the detector output is provided by a detector device in the form of a detector diode coupled between the R.F. power output and the detector output of the power amplifier. The detector output is coupled to a negative input of the error amplifier and an output of the error amplifier is connected to the control input of the power amplifier. Thus the power amplifier, the detector and the error amplifier form a power control loop with negative feedback. The error amplifier also has a positive input to which a reference signal is applied.

[0005] The R.F. output power control described in US 5,378,996 is based on the measurement of the R.F. output power. However, in principle R.F. output power control could also be based on a measurement of the R.F. power amplifier current, i.e. on the power or current consumption of the R.F. power amplifier.

[0006] R.F. output power control based on a measurement of the R.F. power amplifier current is described in Ashok Bindra, "Smart biasing keeps R.F. power amplifier on track", electronic design, 21 January 2002, pp. 38, 40. In this article a monolithic controller that regulates and controls the output power of an R.F. power amplifier is described.

[0007] The controller is part of a closed loop solution that permits calibration of the R.F. power amplifier's gate biasing voltage in real-time modes. A schematic block diagram of an R.F. power amplifier circuit comprising the known controller is depicted in Fig. 1a. As becomes apparent from Fig. 1a, the R.F. power amplifier circuit 10 comprises a power control loop 12 and a signal supply branch 14. The power control loop 12 comprises an R.F. power amplifier 22, a current sensing element in the form of a resistor 24, a detector in the form of a comparator 16 and a filter 18.

[0008] The resistor 24 is used to sense the drain current of the R.F. power amplifier 22. The drain current is converted into a voltage that is fed together with an external voltage reference from the signal supply branch 14 to the comparator 16.

[0009] The output signal of the comparator 16 is filtered by the filter 18 and the filtered signal is used to control the R.F. power amplifier's 22 gate biasing voltage.

[0010] To cope with temperature drift and aging that affect efficiency and linearization of the R.F. power amplifier 22, a control input 14' is provided outside the power control loop 12 between the signal supply branch 14 and the comparator 16. By means of a control signal applied to the control input 14' the output power Pour of the power amplifier 22 can be controlled.

[0011] Document US 6,351,189 B1 concerns a system and a method for auto-bias of an amplifier. The system monitors a physical quantity indicative of the operating state of the amplifier and controls the amplifier bias so as to control the amplifier operating point sufficiently to compensate for variations in amplifier electrical characteristics, amplifier load, amplifier temperature and input signals.

[0012] Document EP 0 905 883 A2 concerns a temperature compensation circuit for field effect transistor amplifier circuits. In one embodiment, an amplifier circuit comprising a power control loop is disclosed. The amplifier has a power control input and a power supply input. A differential amplifier provides an input signal to the amplifier.

[0013] Document WO 03/073603 A2, which is prior art according to Art. 54 (3) EPC for the present invention, concerns a current modulator with a dynamic amplifier impedance compensation. The modulator comprises an amplitude modulation circuit, which provides a modulated supply current to a radio frequency power amplifier. The modulation circuit includes a detection circuit responsive to changes in the ratio of a supply voltage to the modulated supply current.

[0014] Departing from an R.F. output power control scheme taking into account the R.F. power amplifier current, there is a need for a R.F. power amplifier circuit which allows a robust implementation of a power control scheme. There is also a need for a method of controlling the R.F. power amplifier of such an R.F. power amplifier circuit.

SUMMARY OF THE INVENTION

[0015] The above problems are solved by an R.F. power amplifier circuit having the features of claim 1 and a method of controlling the output power of an R.F. power amplifier having the features of claim 9.

[0016] According to the present invention an R.F. power amplifier circuit is provided which comprises a power control loop including an R.F. power amplifier having a power control input and a power supply input, and at least one variable loop element coupled between the power control input and the power supply input of the power amplifier, the at least one variable loop element having a control input configured to reduce variations of control loop parameters.

[0017] The control input of the variable loop element thus enables to actually control the control loop by taking into account a feedback signal characteristic of a current consumption of the power amplifier. This control of the control loop preferably aims at indirectly controlling the output power by attaining a stationary state, whereas prior art solutions aim at changing such a stationary state. The variable loop element may be an element that can be tuned continuously or stepwise. The characteristics of the variable loop element arranged in the feedback path may be controlled such that variations of dynamic loop parameters like the loop damping factor or the natural loop frequency are reduced and, ideally, completely compensated.

[0018] It is thus firstly proposed to base the output power control on a feedback signal characteristic of the power amplifier current and secondly to reduce loop parameter variations, many of which are specific to such a feedback mechanism, by providing one or more variable loop elements in the feedback path. The variable loop elements may be actively or passively controlled for example such that the control loop parameters become linearized or stationary.

[0019] Reduced control loop parameter variations lead to an output power control which is more robust. Furthermore, calibration time required e.g. for power-time-template calibration can be reduced especially in the case of power amplifier circuits that are to be operated in multiple frequency bands.

[0020] The variable loop element may be controlled directly or after signal conversion by a signal readily available at the power amplifier circuit and preferably by a signal related to the output power control like an externally provided reference power control signal fed to the power control loop or a power control signal created within the power control loop. Additionally or alternatively, the variable loop element may be controlled by a dedicated control signal like an offset signal. Preferably, the variable loop element is configured such that it is simultaneously controlled by a readily available signal related to the output power control and a signal related to the output power control and a dedicated control signal.

[0021] The variable loop element has a control input to which a dedicated control signal and/or a readily available but, if required, additionally processed control signal may be fed. This control input allows for example to tune the variable loop element continuously or discretely. In particular, the control input allows to create a further (internal) feedback path (i.e. an internal control loop for the variable loop element) by coupling the control input for example to a particular node of the (external) feedback path between the power control input and the power supply input of the power amplifier. Alternatively, the internal feedback path may be created by coupling the control input of the variable loop element to a node outside the external feedback path. For example the control input may be coupled to a signal supply branch of the power control loop. By means of the internal control loop a feedback signal tapped from the power control loop or the signal supply branch may thus be fed directly or after signal conversion to the control input of the variable loop element.

[0022] The control input of the variable loop element may be coupled to a signal converter which may be arranged in the internal feed back path and which may comprise at least one of a filter circuit, a multiplier, a level shifter, a buffer, a limiter, a look-up table and a voltage or current source. Preferably, the signal converter converts a converter input signal into a converter output signal that is coupled via the control input to the variable loop element.

[0023] The converter input signal is preferably a readily available power control signal or a signal derived therefrom. The signal converter may have his own control terminal to which for example the power control signal or the signal derived therefrom is fed. Alternatively, such a signal may be coupled directly to the control input of the variable loop element. A digital control interface may be coupled either to the control terminal of the signal converter or directly to the control input of the variable loop element. The digita control interface is preferably arranged in the internal feed back path.

[0024] In a preferred embodiment the power amplifier is operable in multiple frequency bands. In such a case the variable loop element may be controlled in each frequency band differently. Such an individual control is preferably performed such that identical loop parameters for all frequency bands are achieved. To that end, frequency band specific control signals may be fed to the variable control element. Identical loop parameters for all frequency bands allow to expedite calibration since calibration values found for one frequency band can be used (in conjunction with the appropriate control signal) for the remaining frequency bands as well.

[0025] Variations of control loop parameters are caused by a plurality of mechanisms. In the case the output power control is based on a feedback signal characteristic of the power amplifier current, variations of the power amplifier constant are a major contribution to loop parameter variations. The power amplifier constant describes the relationship between current consumption and control voltage of the power amplifier.

[0026] It is advantageous if the characteristic of the variable loop element is selected to vary and/or is varied in such a manner that the adverse effects of variations of the power amplifier constant are reduced. Of course, the characteristic of the variable loop element may also vary or be varied such that additional effects or other effects that cause loop parameter variations are reduced.

[0027] The variable loop element may be a dedicated component arranged in the feedback path solely for the purpose of reducing loop parameter variations. Additionally or alternatively, a component already present in the feed path, for example a filter, a sensing element or a detector, may be configured such that the component allows in addition to its primary task a deliberate reduction of loop parameter variations.

[0028] Preferably, the variable loop element is constituted by a variable filter like a loop filter or a low pass filter of the power control loop. The variable filter may comprise at least one of a variable resistor and a variable capacitance. Furthermore, in the case of an active filter a dedicated control routine may be implemented which allows to reduce loop parameter variations.

[0029] The variable loop element may be constituted by or may comprise a varicap diode. Such a varicap diode provides a variable capacitance which is controlled by the voltage across the anode terminal and the cathode terminal. Thus the varicap diode may be controlled by the loop control signal and more particularly by the loop control voltage. An additional control signal may be applied to either one or both of the two terminals of the varicap diode to introduce a further control parameter.

[0030] If used in a filter arrangement, the varicap diode renders the filter variable. However, the varicap diode may also be used in conjunction with other variable loop elements like a loop detector. Instead of or in addition to a varicap diode, the loop detector may have a variable gain which is controlled such that loop parameter variations are reduced.

[0031] The invention described above may be implemented as a hardware solution or as a software solution. In the case of a software solution the invention may be realized in the form of a computer program product comprising program code portion for performing the steps of the invention. This computer program product may be stored on a computer readable recording medium.

[0032] According to a preferred embodiment of the invention, the R.F. power amplifier circuit of the invention is arranged in a network component like a mobile terminal for wireless communications (for example a multi-band mobile telephone) or a driver stage of a base station of a mobile communications network.

BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Further aspects and advantages of the invention will become apparent upon reading the following detailed description of preferred embodiments of the invention and upon reference to the drawings, in which:

Fig. 1a

is a block diagram of a prior art R.F. power amplifier circuit;

Fig. 1b

is a block diagram of an R.F. power amplifier circuit according to the invention;

Fig. 2

is a diagram depicting the relationship between the power amplifier constant and the power amplifier control voltage;

Fig. 3

shows a first implementation of a static loop filter;

Fig. 4

shows a second implementation of a static loop filter;

Fig. 5a and 5b

schematically show a first embodiment in accordance with the invention of a variable loop element in the form of a variable loop filter;

Fig. 6

schematically shows a second embodiment in accordance with the invention of a variable loop element in the form of a variable loop filter;

Fig. 7

shows the relationships between the power amplifier control voltage and the variable capacitances of the two embodiments depicted in Figs. 5 and 6;

Fig. 8

shows the variation of the damping factor in dependence of the power amplifier control voltage for a prior art loop filter capacitor and a variable loop filter capacitor according to the invention;

Fig. 9

schematically shows a third embodiment in accordance with the invention of a variable loop element in the form of a variable loop filter;

Fig. 10

schematically shows a fourth embodiment in accordance with the invention of a variable loop element in the form of a variable loop filter;

Fig. 11

schematically shows a fifth embodiment in accordance with the invention of a variable loop element in the form of a variable loop filter;

Fig. 12

shows a block diagram of an R.F. power amplifier circuit comprising the variable loop element depicted in Fig. 11 or 12;

Fig. 13

shows a practical arrangement of a variable loop filter in a variable loop filter;

Fig. 14

shows the relationship between the power amplifier output power and the power amplifier control voltage for the arrangement depicted in Fig. 13; and

Figs. 15a and 15b

show the power amplifier output power as a function of time step response of a prior art power amplifier circuit and of a power amplifier circuit according to the invention.

DESCRIPTION OF PREFERRED EMBODIMENTS

[0034] In the following the invention is exemplarily set forth with respect to an R.F. power amplifier circuit comprising a variable loop element in the form of a variable loop filter. It should be noted however that in principle any other element of the power control loop could be modified such that it functions as a variable loop element in addition to or instead of the variable loop filter. Moreover, although the invention is exemplarily explained in conjunction with a variable loop capacitor, other variable components like variable resistors and other variable parameters like a variable detector gain may be used also to implement a variable loop element.

[0035] Furthermore, the following discussion of the preferred embodiments does not take temperature drift of the variable capacitors or requirements for properly biasing the variable capacitors into consideration. In practical realizations, appropriate means to compensate for the temperature drift and means to properly bias the variable capacitors will have to be provided.

[0036] In Fig. 1b block diagram of an exemplary R.F. power amplifier circuit 10 according to the invention is shown. The power amplifier circuit 10 includes a power control loop 12 and a reference voltage supply branch 14 coupled to the power control loop 12.

[0037] The power control loop 12 comprises a detector circuit 16, a loop filter 18, an optional low pass filter 20, a power amplifier 22 and a current sensing element 24. In the exemplary embodiment described with reference to Fig. 1b, the detector circuit 16 comprises an error amplifier having a negative input 26 and a positive input 28. The positive input 28 is coupled to the reference voltage supply branch 14 which includes a first pulse shaping filter 32 in communication with the positive input 28 of the detector circuit 16, an exponential amplifier 30 having an output coupled to the first pulse shaping filter 32, and a second pulse shaping filter 28 coupled to the input of the exponential amplifier 30.

[0038] An output of the detector circuit 16 is coupled to the loop filter 18 which has a control input 46 configured to reduce loop parameter variations. The loop filer 18 is coupled via the low pass filter 20 to a power control input 34 of the power amplifier 22. The power amplifier 22 further has a power supply input 36, an R.F. power output 38 and an R.F. signal input not shown in Fig. 1b. The current sensing element 24 is constituted by a resistor that is coupled between the power supply input 36 of the power amplifier 22 and a current supply (V_bat).

[0039] The basic operation of the power amplifier circuit 10 depicted in Fig. 1b is as follows.

[0040] A discrete control voltage POWLEV* fed into the second pulse shaping filter 28 is converted in the reference voltage supply branch 14 into a continuous power amplifier reference voltage POWLEV that is applied to the positive input 28 of the detector circuit 16. In the detector circuit 16 this power amplifier reference voltage POWLEV is compared with a feedback signal generated by the current sensing element 24. The difference signal is amplified by the detector circuit 16 and fed in the form of a power amplifier control voltage PAREG via the loop filter 18 and the low pass filter 20 to the power control input 34 of the power amplifier 22. The power amplifier 22 amplifies an R.F. input signal in accordance with the power amplifier control voltage PAREG and outputs the amplified signal via its power output 38.

[0041] As can be seen from Fig. 1b, the power control loop 12 comprises a feedback path coupled between the power control input 34 and the power supply input 36 of the power amplifier 22. This feedback path comprises the current sensing element 24, the detector circuit 16, the loop filter 18 and the low pass filter 20.

[0042] By coupling the feedback path to the power supply input 36 of the power amplifier 22, the output power control of the power amplifier 22 is based on a feedback signal characteristic of the current consumption of the power amplifier 22. As a result of the fact that output power control is based on a feedback signal characteristic of the current consumption of the power amplifier 22, the power amplifier constant K_pa, which describes the relationship between current consumption and control voltage PAREG of the power amplifier 22, influences the dynamic parameters of the power control loop 12. This becomes apparent from the transfer function H(s) of the power control loop 12, which can be written


where the transfer function Hf(s) of the loop filter 18 can exemplarily be written as

the transfer function Hlp(s) of the low pass filter 20 can be written as

the transfer function K_sense of the current sensing element 24 can be written as

and the transfer function K_detector of the detector circuit 16 can be written as

[0043] In the following, two specific control loop parameters, namely the loop damping factor d and the natural loop frequency w_n, of the power control loop 12 will exemplarily be considered in more detail. These control loop parameters can be written as

and

[0044] From equations (6) and (7) it is obvious that dynamic properties like the loop damping factor d and the loop bandwidth of the power control loop 12 strongly depend on variations of the power amplifier constant K_pa. The power amplifier constant K_pa, however, strongly varies with the power amplifier control voltage PAREG. This becomes apparent from Fig. 2 that shows a diagram depicting the functional relationship between the power amplifier constant K_pa and the power amplifier control voltage PAREG as derived on the basis of a power amplifier model.

[0045] It has been experimentally found that the power amplifier constant K_pa of a typical power amplifier for the 900 MHz band of the Global System for Mobile communications (GSM 900) varies between 1,6 A/V and 0,96 A/V, that the power amplifier constant K_pa of a typical triple band GSM900/GSM1800/GSM1900 power amplifier ranges between 3A/V and 0,2 A/V, and that a dual band power amplifier has a maximum power amplifier constant K_pa which can be as high as 6 A/V.

[0046] As becomes apparent from the above, the relationship between the power amplifier control voltage PAREG and the power amplifier constant K_pa is highly non-linear. Consequently, typical dynamic control loop parameters like the loop damping factor d and the natural loop frequency W_n strongly vary with the power amplifier control voltage PAREG.

[0047] Such variations of the control loop parameters render power control loop design very difficult. Power control loop design has to ensure that on the one hand the loop bandwidth is wider than the bandwidth of the pulse shaping filters 28, 32 to ensure that pulse shaping remains independent of power control loop parameters. On the other hand the loop bandwidth shall be as small as possible (with constant damping factor) in order to reduce noise. Furthermore, constant control loop parameters are advantageous from a calibration point of view. Power control loop design aims at finding a compromise on all aspects discussed above. This requires, however, that variations of the control loop parameters are reduced as strong as possible.

[0048] According to the exemplary embodiment of the invention discussed in context with Fig. 1b, a variable loop element with a control input is provided which reduces the loop parameter variations that result from variations of the power amplifier constant K_pa. Of course, variations of additional parameters or of other parameters apart from the power amplifier constant K_pa could also be compensated in accordance with the invention to linearize the control loop parameters.

[0049] In principle, variations of the power amplifier constant K_pa in equations (6) and (7) could be compensated by rendering one or more of the other parameters of equations (6) and (7) variable and be varying these one or more other parameters appropriately. This means that basically a variable loop element in the form of a variable current sensing element 24, a variable detector circuit 16, a variable loop filter 18 and/or a variable low pass filter 20 could be provided. Additionally or alternatively, a dedicated variable loop element could be introduced into the power control loop 12.

[0050] As becomes apparent from equations (6) and (7), the capacitance C_c of the loop filter 18 and the gain G_m of the detector circuit 16 are parameters that can especially advantageously be used for compensating variations of control loop parameters that are induced by a varying power amplifier constant K_pa. In the following, linearization of the power control loop 12 is exemplarily illustrated in conjunction with a variable capacitance C_c.

[0051] Departing from the relationship between the power amplifier constant K_pa and the control voltage PAREG depicted in Fig. 2 on the one hand and typical values for the individual parameters of equations (6) and (7) on the other hand, the damping factor d and the natural loop frequency w_n can vary as illustrated in the following table:

Gm	0.008 S	0.008 S
Cc	100 pF	100 pF
Rlp	56 Ω	56 Ω
Clp	1nF	1nF
Rsense	0.05 Ω	0.05 Ω
Kpa	2.84 A/V (PAREG=1.7V)	0.03 A/V (PAREG=2.5V)
d	2.0	22.3
wn	4.501 x 106rad/s	4.401 x 106rad

[0052] The above variation of the damping factor d and of the natural loop frequency w_n is made less dependent on the variation of the power amplifier constant K_pa by introducing a variable loop filter having a variable capacitance C_c that can be varied at the same rate as the variation of K_pa. However, prior to discussing realization of variable loop filters, possible implementations of (static) realizations are considered with reference to Figs. 3 and 4.

[0053] According to a first variant, the detector circuit is realized in the form of an amplifier as part of a mixed signal ASIC, whereas the passive loop filter is created with a discrete capacitor C and a discrete resistor R' which form a PI loop filter as depicted in Fig. 3. As becomes apparent from Fig. 3, the discrete capacitor C is connected single sided to ground.

[0054] The transfer function Hf(s) of the loop filter depicted in Fig. 3 can be written as follows:

[0055] Equation (8) basically corresponds to equation (2) and equations (6) and (7) could be rewritten accordingly.

[0056] According to a second variant, the detector circuit and the loop filter might be realized as depicted in Fig. 4, i.e. with an amplifier stage 40 and a driver stage 42. The amplifier stage 40 and the driver stage 42 are both part of a mixed signal ASIC, whereas loop filter capacitor C remains a discrete component. The transfer function F(s) of the combined amplifier stage 40 and driver stage 42 depicted in Fig. 4 can be written as

[0057] The implementation depicted in Fig. 4 is advantageous because compared to the implementation depicted in Fig. 3 there is no external resistor required for the loop filter.

[0058] In Fig. 5a a first variant of a variable loop filter 18 according to the invention is depicted. As becomes apparent from Fig. 5a, the variable loop filter 18 comprises a resistor R' and a variable loop capacitor C_var in the form of a varicap diode. The loop capacitor C_var depicted in Fig. 5a is the equivalent of loop capacitor C depicted in Fig. 3, but the capacitance of loop capacitor C_var is variable and basically controlled by the power amplifier control voltage PAREG, i.e. changes approximately at the same rate as the power amplifier constant K_pa varies.

[0059] The variable loop filter 18 has a control input 46 that is coupled between a cathode of the variable loop capacitor C_var on the one hand and a common node to which the resistor R' is coupled on the other hand. An external resistor R'' is also coupled to this common node and provides a control signal from the detector 16 to the control input 46 of the variable loop filter 18. Thus an internal control loop including the variable loop capacitor C_var is formed. An internal control signal of the control loop 12 is tapped via the resistor R'' and fed to the control input 46 coupled to the variable loop capacitor C_var.

[0060] In principle, the two parallel resistors R', R'' depicted in Fig. 5a could be combined to a single resistor R. A corresponding equivalent circuit of the variable loop filter 18 of Fig. 5a is depicted in Fig. 5b.

[0061] The loop filter 18 depicted in Figs. 5a and 5b constitutes a passive PI loop filter with variable "I" part due to the variable capacitor C_var. The transfer function Hf(s) of the loop filter 18 becomes a function of the voltage U_cvar across the terminals of C_var.This allows to reduce variations of control loop parameters that are caused by variations of the power amplifier constant K_pa.

[0062] Fig. 6 shows the same PI loop filter configuration as depicted in Figs. 5a and 5b but with an (additional) control terminal 46 for the loop capacitor C_var that is coupled between an anode of the loop capacitor C_var and an additional capacitor C_o which provides a dc block to ground. A control signal in the form of a control voltage U_offset may be applied to the control input 46 of the variable loop filter 18. In the presence of U_offset the value of C_var__offset is determined by the difference between the power amplifier control voltage PAREG and the control voltage U_offset. The control voltage U_offset thus allows to tune the loop capacitor C_var in order to even better reduce variations of control loop parameters.

[0063] The transfer function Hf(s) of the loop filter 18 depicted in Fig. 6, which is a function of the voltage difference U_cvar between the power amplifier control voltage PAREG and the control voltage U_offset, can be written as

with

[0064] As a result of the control voltage U_offest applied to the control input 46 of the variable loop filter 18, the C_var__offset versus power amplifier control voltage PAREG curve is shifted along the x-axis. Such a tuning is extremely useful for matching the characteristics of C_var__offset to the characteristics of the power amplifier constant K_pa.

[0065] The characteristics of the C_var and C_var__offset are depicted in Fig. 7. As becomes apparent from Fig. 7, C_{var_offset} is the shifted replica of C_var.

[0066] In Fig. 8 a comparison of the damping factors d for a loop filter having a fixed capacitance C and a variable loop filter 18 as depicted in Fig. 6 having a tuned and variable loop filter capacity C_{var_offset} is shown. The use of the variable loop filter 18 allows to reduce the variation of the damping factor d by approximately a factor of 2. This also becomes apparent from the table below.

	Cc=Cvar_offset	Cc=const
Kpa	2.84 A/V (PAREG=1.7V)
D	1.6	2.0
Wn	5.601 x 106rad/s	9.501 x 106rad/s
Kpa	0.03 A/V (PAREG=2.5V)
D	10.5	22.3
Wn	0.852 x 106rad/s	0.401 x 106rad/s
d_ratio	6.6	11.2
wn_ratio	0.152	0.089

[0067] Further approvements can be achieved by using a varicap diode with a capacitance characteristic that better matches the characteristic of the power amplifier constant K_pa and by actively controlling the control voltage U_offset.

[0068] Since the power amplifier constant K_pa also varies with frequency, the characteristics of the power amplifier constant K_pa will be different for different frequency bands. The control voltage U_offset can thus be used to tune C_{var_offset} for each frequency band individually to achieve identical control loop parameters for all frequency bands. In the case of identical control loop parameters power-time-template calibration for multiple frequency band mobile telephones is expedited because calibration values found for one frequency band can readily be used (if the appropriate control voltage U_offset is applied) for the remaining frequency bands as well. Consequently, the calibration time might be reduced by more than 50 % for a triple band mobile telephone.

[0069] An additional resistor R_c could be added, for example for filtering purposes, to the variable loop filter 18 as shown in Fig. 9. The resistor R_c is coupled to the control terminal 46 and a modified control voltage U_offset* has to be applied to the resistor R_c.

[0070] As depicted in Fig. 10, the control signal U_offset* can be provided by a signal converter 50 which helps to better adapt the characteristic of C_var__offset of the varicap diode to the characteristic of the power amplifier constant K_pa. The signal converter 50 comprises a control terminal 52 for receiving a control signal U_offset** and an output terminal 54 coupled to the resistor R_c. In principle, the resistor R_c depicted in Fig. 10 could be omitted and the output terminal 54 of the signal converter 50 could be directly coupled to the control input 46.

[0071] The signal converter 50 linearily or non-linearily transforms the control signal U_offset** into the control signal U_offset** in accordance with the relationship U_offset* = f_converter(U_offset**). The signal converter 50 may comprise a filter circuit, a multiplier, a level shifter, a buffer, a limiter, a look-up table or a voltage/current source.

[0072] An alternative embodiment of the variable loop filter 18 with C_var__offset connected single sided to ground is depicted in Fig. 11.

[0073] In principle, the control input 46 of the variable loop filter 18 could be connected directly or indirectly to the power amplifier control voltage PAREG or another power control signal like the power amplifier reference voltages POWLEV or POWLEV*. In this regard Fig. 12 exemplarily shows the block diagram of the power control loop 12 with the control terminal 52 of the signal inverter 50 coupled to the PAREG signal to create an internal control loop 12'. It should be noted that the control input 46 of the variable loop filter 18, or of any other variable loop element (for example the current sensing element 24 or the detector 16), could alternatively be coupled to a node arranged between the pulse shaping filter 28 and the detector 16 or the detector 16 and the loop filter 18. Of course, U_offset, U_offset* or U_offset** could also be supplied directly from a digital control interface like an analog/digital converter.

[0074] Fig. 13 shows a circuit with a variable loop filter used for practical measurements performed on a mobile telephone board. The circuit of Fig. 13 is based on the circuit of Fig. 4 with the static loop filter.

[0075] Returning to Fig. 13, resistor R_d is used to provide a dc path for the varicap diode C_var. Capacity C_dc is used to provide dc decoupling for the signal U_offset**. The sigal U_offset** is supplied by an external, variable voltage source. The diagram of Fig. 14 shows the power amplifier output power versus power control voltage POWLEV as measured for the circuit of Fig. 13. From Fig. 14 it becomes apparent that the loop filter arrangement depicted in Fig. 13 helps to reduce the overshot in output power compared to the unmodified circuit depicted in Fig. 4.

[0076] Figs. 15a and 15b show the output power versus time step response of the unmodified and modified circuit depicted in Figs. 4 and 13, respectively. It can be clearly seen that the maximum power overshot is reduced by about more than 75 %.

Claims

1. An R.F. power amplifier circuit (10) comprising a power control loop (12) with an R.F. power amplifier (22) having a power control input (34) and a power supply input (36) and a feedback path, the feedback path being coupled between the power control input (34) and the power supply input (36) and comprising a current sensing element (24), a detector circuit (16), and a loop filter (18), the current sensing element (24) being configured to generate a feedback signal from the power amplifier (22) and to supply the feedback signal to the detector circuit (16), the detector circuit (16) being configured to compare the feedback signal with a power amplifier reference voltage (POWLEV) in order to obtain a difference signal, to amplify the difference signal, and to supply the amplified difference signal via the loop filter (18) as a power amplifier control voltage (PAREG) to the power control input (34) of the R.F. power amplifier (22), wherein
at least one of the loop filter (18) and the current sensing element (24) has variable characteristics, and wherein
the at least one of the elements having variable characteristics (18, 24) has a control input (46) to reduce variations of control loop parameters (d, w_n).

2. The power amplifier circuit of claim 1,
wherein the control input (46) is coupled to the power control loop (12) or to a signal supply branch (14) of the power control loop (12) to form an internal control loop (12') which includes the at least one element having variable characteristics (18, 24).

3. The power amplifier circuit of claim 1 or 2,
wherein the at least one element having variable characteristics (18, 24) is a variable filter (18) or a variable current sensing element (24).

4. The power amplifier circuit of claim 3,
wherein the variable filter (18) comprises at least one of a variable resistor and a variable capacitance (C_var).

5. The power amplifier circuit of one of claims 1 to 4,
wherein the at least one element having variable characteristics (18, 24) comprises a varicap diode (C_var).

6. The power amplifier circuit of one of claims 1 to 5,
further comprising a signal converter (50) coupled to the control input (46).

7. The power amplifier circuit of one of claims 1 to 6,
wherein the control input (46) or a control terminal (52) of the signal converter (50) is coupled to a digital control interface.

8. A component of a wireless communications network comprising the power amplifier circuit (10) of one of claims 1 to 7.

9. A method of controlling the output power of an R.F. power amplifier (22) provided within a power control loop (12), the power control loop (12) comprising a feedback path coupled between a power control input (34) and a power supply input (36) of the R.F. power amplifier (22), the feedback path comprising a current sensing element (24), a detector circuit (16), and a loop filter (18), wherein the current sensing element (24) generates a feedback signal from the power amplifier (22) and supplies the feedback signal to the detector circuit (16), the detector circuit (16) compares the feedback signal with a power amplifier reference voltage (POWLEV) in order to obtain a difference signal, amplifies the difference signal, and supplies the amplified difference signal via the loop filter (18) as a power amplifier control voltage (PAREG) to the power control input (34) of the R.F. power amplifier (22), wherein at least one of the loop filter (18) and the current sensing element (24) has variable characteristics, wherein the at least one element having variable characteristics (18, 24) has a control input (46), and wherein the characteristics of the at least one element having variable characteristics (18, 24) are varied by applying a control signal to the control input (46) such that variations of control loop parameters (d, w_n) are reduced.

10. The method of claim 9,
wherein the characteristics are varied to reduce variations of control loop parameters (d, w_n) that result from variations of the power amplifier constant (K_pa).

11. The method of claim 9 or 10,
wherein a feedback signal tapped from the power control loop (12) or from a signal supply branch (14) of the power control loop (12) is directly or after signal conversion fed to the control input (46).

12. The method of one of claims 9 to 11,
wherein the characteristics are controlled by a power control signal (POWLEV, PAREG) for the power amplifier (22) or a signal derived therefrom.

13. The method of one of claims 9 to 12,
wherein the characteristics are controlled by a dedicated control signal.

14. The method of one of claims 9 to 13,
wherein the R.F. power amplifier (22) is operable in multiple frequency bands and
wherein the characteristics are individually controlled in each frequency band.

15. A computer program product comprising program code portions for performing the steps of at least one of claims 9 to 14.

16. The computer program product of claim 15, stored on a computer readable recording medium.

Ansprüche

1. RF-Leistungsverstärkerschaltung (10), umfassend eine Leistungs-Regelschleife (12) mit einem RF-Leistungsverstärker (22), welcher einen Leistungssteuereingang (34) und einen Stromversorgungseingang (36) und einen Rückkopplungsweg aufweist, wobei der Rückkopplungsweg zwischen den Leistungssteuereingang (34) und den Stromversorgungseingang (36) gekoppelt ist und ein Strommesselement (24), eine Detektorschaltung (16) und ein Schleifenfilter (18) umfasst, wobei das Strommesselement (24) dazu konfiguriert ist, ein Rückkopplungssignal vom Leistungsverstärker (22) zu erzeugen und das Rückkopplungssignal der Detektorschaltung (16) zuzuführen, wobei die Detektorschaltung (16) konfiguriert ist, das Rückkopplungssignal mit einer Leistungsverstärker-Referenzspannung (POWLEV) zu vergleichen, um ein Differenzsignal zu erhalten, das Differenzsignal zu verstärken und das verstärkte Differenzsignal durch das Schleifenfilter (18) als eine Leistungsverstärker-Steuerspannung (PAREG) dem Leistungssteuereingang (34) des RF-Leistungsverstärkers (22) zuzuführen, wobei
das Schleifenfilter (18) und/oder das Strommesselement (24) variable Charakteristika aufweist und wobei
das wenigstens eine der Elemente, welches variable Charakteristika (18, 24) aufweist, einen Steuereingang hat (46), um Schwankungen von Parametern (d, W_n) der Regelschleife zu reduzieren.

2. Leistungsverstärkerschaltung nach Anspruch 1,
wobei der Steuereingang (46) mit der Leistungs-Regelschleife (12) oder mit einem Signal-Versorgungszweig (14) der Leistungs-Regelschleife (12) gekoppelt ist, um eine interne Regelschleife (12') auszubilden, welche das wenigstens eine Element enthält, welches variable Charakteristika (18, 24) aufweist.

3. Leistungsverstärkerschaltung nach Anspruch 1 oder 2,
wobei das wenigstens eine Element, welches variable Charakteristika (18, 24) aufweist, ein variables Filter (18) oder ein variables Strommesselement (24) ist.

4. Leistungsverstärkerschaltung nach Anspruch 3,
wobei das variable Filter (18) einen variablen Widerstand und/oder eine variable Kapazität (C_var) umfasst.

5. Leistungsverstärkerschaltung nach einem der Ansprüche 1 bis 4, wobei das wenigstens eine Element, welches variable Charakteristika (18, 24) aufweist, eine Kapazitätsdiode (C_var) umfasst.

6. Leistungsverstärkerschaltung nach einem der Ansprüche 1 bis 5, zusätzlich umfassend einen Signalwandler (50), der mit dem Steuereingang (46) gekoppelt ist.

7. Leistungsverstärkerschaltung nach einem der Ansprüche 1 bis 6, wobei der Steuereingang (46) oder ein Steueranschluss (52) des Signalwandlers (50) mit einer digitalen Steuerschnittstelle gekoppelt ist.

8. Komponente eines drahtlosen Kommunikationsnetzes, umfassend die Leistungsverstärkerschaltung (10) nach einem der Ansprüche 1 bis 7.

9. Verfahren zum Steuern der Ausgangsleistung eines RF-Leistungsverstärkers (22), welcher innerhalb einer Leistungs-Regelschleife (12) bereitgestellt wird, wobei die Leistungs-Regelschleife (12) einen Rückkopplungsweg umfasst, der zwischen einen Leistungssteuereingang (34) und einen Stromversorgungseingang (36) des RF-Leistungsverstärkers (22) gekoppelt ist, wobei der Rückkopplungsweg ein Strommesselement (24), eine Detektorschaltung (16) und ein Schleifenfilter (18) umfasst, wobei das Strommesselement (24) ein Rückkopplungssignal vom Leistungsverstärker (22) erzeugt und das Rückkopplungssignal der Detektorschaltung (16) zuführt, wobei die Detektorschaltung (16) das Rückkopplungssignal mit einer Leistungsverstärker-Referenzspannung (POWLEV) vergleicht, um ein Differenzsignal zu erhalten, das Differenzsignal verstärkt und das verstärkte Differenzsignal durch das Schleifenfilter (18) als eine Leistungsverstärker-Steuerspannung (PAREG) dem Leistungssteuereingang (34) des RF-Leistungsverstärker (22) zuführt, wobei das Schleifenfilter (18) und/oder das Strommesselement (24) variable Charakteristika aufweist, wobei das wenigstens eine Element, welches variable Charakteristika (18, 24) aufweist, einen Steuereingang (46) hat und wobei die Charakteristika des wenigstens einen Elements, welches variable Charakteristika (18, 24) aufweist, durch Anlegen eines Steuersignals an den Steuereingang (46) variiert werden, so dass Schwankungen von Parametern (d, w_n) der Regelschleife reduziert werden.

10. Verfahren nach Anspruch 9, wobei die Charakteristika variiert werden, um Schwankungen von Parametern (d, w_n) der Regelschleife zu reduzieren, welche aus Schwankungen der Leistungsverstärker-Konstante (K_pa) resultieren.

11. Verfahren nach Anspruch 9 oder 10, wobei ein Rückkopplungssignal, welches von der Leistungs-Regelschleife (12) oder von einem Signalversorgungszweig (14) der Leistungs-Regelschleife (12) abgegriffen wird, direkt oder nach Signalumwandlung dem Steuereingang (46) eingespeist wird.

12. Verfahren nach einem der Ansprüche 9 bis 11, wobei die Charakteristika durch ein Leistungssteuersignal (POWLEV, PAREG) für den Leistungsverstärker (22) oder ein davon abgeleitetes Signal gesteuert werden.

13. Verfahren nach einem der Ansprüche 9 bis 12, wobei die Charakteristika durch ein dediziertes Steuersignal gesteuert werden.

14. Verfahren nach einem der Ansprüche 9 bis 13, wobei der RF-Leistungsverstärker (22) in mehreren Frequenzbändern betreibbar ist, und wobei die Charakteristika in jedem Frequenzband individuell gesteuert werden.

15. Computerprogrammprodukt, welches Teile eines Programmcodes für das Durchführen der Schritte wenigstens eines der Ansprüche 9 bis 14 umfasst.

16. Computerprogrammprodukt nach Anspruch 15, welches auf einem Computerlesbaren Aufzeichnungsmedium gespeichert ist.

Revendications

1. Circuit (10) d'amplification de puissance R.F comprenant une boucle (12) de commande de puissance avec un amplificateur (22) de puissance R.F ayant une entrée (34) de commande de puissance et une entrée (36) d'alimentation de puissance et une voie de rétroaction, la voie de rétroaction étant couplée entre l'entrée (34) de commande de puissance et l'entrée (36) d'alimentation de puissance et comprenant un élément (24) de détection de courant, un circuit de détection (16), et un filtre de boucle (18), l'élément (24) de détection de courant étant configuré pour générer un signal de rétroaction à partir de l'amplificateur (22) de puissance et pour alimenter le signal de rétroaction au circuit de détection (16), le circuit de détection (16) étant configuré pour comparer le signal de rétroaction à une tension de référence (POWLEV) de l'amplificateur de puissance afin d'obtenir un signal de différence, pour amplifier le signal de différence, et pour alimenter le signal de différence amplifié à travers le filtre de boucle (18) en tant que tension de commande (PAREG) de l'amplificateur de puissance à l'entrée (34) de commande de puissance de l'amplificateur (22) de puissance R.F, dans lequel
au moins l'un parmi le filtre de boucle (18) et l'élément (24) de détection de courant possède des caractéristiques variables, et dans lequel
l'au moins un des éléments ayant des caractéristiques variables (18, 24) a une entrée de commande (46) afin de réduire des variations de paramètres (d, w_n) de boucle de commande.

2. Circuit d'amplification de puissance de la revendication 1,
dans lequel l'entrée de commande (46) est couplée à la boucle (12) de commande de puissance ou à une branche (14) d'alimentation de signal de la boucle (12) de commande de puissance afin de former une boucle (12') de commande interne qui comporte l'au moins un élément ayant des caractéristiques variables (18, 24).

3. Circuit d'amplification de puissance de la revendication 1 ou 2,
dans lequel l'au moins un élément ayant des caractéristiques variables (18, 24) est un filtre (18) variable ou un élément (24) de détection de courant variable.

4. Circuit d'amplification de puissance de la revendication 3,
dans lequel le filtre (18) variable comprend au moins l'une parmi une résistance variable et une capacitance variable (C_var).

5. Circuit d'amplification de puissance de l'une des revendications 1 à 4,
dans lequel l'au moins un élément ayant des caractéristiques variables (18, 24) comprend un varacteur (C_var).

6. Circuit d'amplification de puissance de l'une des revendications 1 à 5,
comprenant en plus un convertisseur de signal (50) couplé à l'entrée de commande (46).

7. Circuit d'amplification de puissance de l'une des revendications 1 à 6,
dans lequel l'entrée de commande (46) ou une borne de commande (52) du convertisseur (50) de signal est couplé à une interface de commande numérique.

8. Composant d'un réseau de communications sans fil comprenant le circuit (10) d'amplification de puissance de l'une des revendications 1 à 7.

9. Procédé de commande de la puissance de sortie d'un amplificateur (22) de puissance R.F prévu dans une boucle (12) de commande de puissance, la boucle (12) de commande de puissance comprenant une voie de rétroaction couplée entre une entrée (34) de commande de puissance et une entrée (36) d'alimentation de puissance de l'amplificateur (22) de puissance R.F, la voie de rétroaction comprenant un élément (24) de détection de courant, un circuit de détection (16), et un filtre de boucle (18), dans lequel l'élément (24) de détection de courant génère un signal de rétroaction à partir de l'amplificateur (22) de puissance et alimente le signal de rétroaction au circuit de détection (16), le circuit de détection (16) compare le signal de rétroaction à une tension de référence (POWLEV) de l'amplificateur de puissance afin d'obtenir un signal de différence, amplifie le signal de différence, et alimente le signal de différence amplifié à travers le filtre de boucle (18) en tant que tension de commande (PAREG) de l'amplificateur de puissance à l'entrée (34) de commande de puissance de l'amplificateur (22) de puissance R.F, dans lequel au moins l'un parmi le filtre de boucle (18) et l'élément (24) de détection de courant électrique possède des caractéristiques variables, dans lequel l'au moins un élément ayant des caractéristiques variables (18, 24) a une entrée de commande (46), et dans lequel on fait varier les caractéristiques de l'au moins un élément ayant des caractéristiques variables (18, 24) en appliquant un signal de commande à l'entrée de commande (46) de telle sorte que des variations de paramètre (d, w_n) de la boucle de commande soient réduites.

10. Procédé de la revendications 9,
dans lequel on fait varier les caractéristiques afin de réduire des variations de paramètres (d, w_n) de la boucle de commande qui résultent de variations de la constante (K_pa) d'amplification de puissance.

11. Procédé de la revendications 9 ou 10,
dans lequel un signal de rétroaction tiré de la boucle (12) de commande de puissance ou d'une branche (14) d'alimentation de signal de la boucle (12) de commande de puissance est alimenté, directement ou après conversion de signal, à l'entrée de commande (46).

12. Procédé de l'une des revendications 9 à 11,
dans lequel les caractéristiques sont commandées par un signal (POWLEV, PAREG) de commande de puissance pour l'amplificateur de puissance (22) ou un signal qui en est dérivé.

13. Procédé de l'une des revendications 9 à 12,
dans lequel les caractéristiques sont commandées par un signal de commande dédié.

14. Procédé de l'une des revendications 9 à 13,
dans lequel l'amplificateur (22) de puissance R.F peut fonctionner dans plusieurs bandes de fréquences et dans lequel les caractéristiques sont commandées individuellement dans chaque bande de fréquences.

15. Produit de programme informatique comprenant des parties de code de programme pour exécuter les étapes an moins de l'une des revendications 9 à 14.

16. Produit de programme informatique de la revendication 15, stocké sur un support d'enregistrement lisible par un ordinateur.

Drawing